Dear DHIS2 Developers/Researchers,
I would like to inquire whether there are ongoing efforts to integrate AI-powered assistants into DHIS2/eLMIS, or if such initiatives are being considered. I believe this could represent a major leap forward for the platform and its users.
The concept is AI-Powered HMIS Early Warning & Decision Support. Instead of DHIS2/eLMIS only showing what has happened, an intelligence layer could help answer:
- What is happening?
- Why is it happening?
- What is likely to happen?
- What action should we consider?
For example, weekly data pulled automatically via API could be analyzed by models and rules to generate risk scores for each governorate, district, or facility. This could highlight:
- Unusual drops in immunization coverage.
- Declining reporting completeness or timeliness.
- Potential data-entry errors.
- Facilities with high vaccine consumption but low immunization numbers.
- Areas at risk of Zero-Dose children.
- Medicines approaching stock-out or expiry.
- Opportunities for redistribution between governorates.
Instead of simply showing Stock-out risk: HIGH, the system could provide actionable intelligence such as:
Aden – Vaccine X
Current stock: 1,850 doses
Average weekly consumption: 620
Estimated stock-out: 3 weeks
Nearby Governorate Y surplus: 2,400 doses
Suggested action: Redistribute 1,000 doses before initiating new supply request.
This transforms DHIS2/eLMIS from data visualization into decision intelligence. Importantly, AI would not replace the system or human decision-making, but act as a supportive layer that explains risks and suggests actions, leaving the final judgment to program managers and ministries of health.
Has such integration already started, or are there plans to explore it? I believe this would add significant value for all stakeholders.
Warm regards,
Sami